JP2004336754A - Method and apparatus for q-factor monitoring using forward error correction coding - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system, method and apparatus for monitoring impairment related parameters such as a Q factor within an all-optical system. <P>SOLUTION: A forward error correction (FEC) is used to derive a bit error rate (BER) and the BER is used to determine the impairment related parameters. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to optical networks, and more particularly to optical performance for monitoring such networks.

  The inability to measure Bit-Error Rate (BER) in the optical layer has long been viewed as a disadvantage of optical networks. BER measurements are considered essential for the management and control of intelligent network elements. Currently, performance monitoring, such as a Synchronous Optical NETwork (SONET) system, is achieved by accessing individual bits recovered in the SONET frame header through the use of an electro-optical interface. The optical signals are converted into electrical signals for analysis and amplification, and then reproduced into optical signals. In this manner, bit-wise processing of overhead bytes is provided to perform network maintenance tasks such as alarm monitoring, Loss Of Frame (LOF) detection, Loss Of Signal (LOS) detection, and the like. Is done.

  While the above technique is useful, the desire to eliminate the electro-optical interface and transmission system in an optical network means that the bit error rate measurement of the optical layer needs to be determined in some way. Bit error rate test sets are not suitable for built-in optical performance monitoring because of their high cost and signal power performance requirements, dispersion tolerance, and fixed data patterns. In addition, the signal measured before reaching the destination is usually error-free, so other measures such as Q-factor must be used to obtain an assessment of BER and signal quality. The Q-factor can be stated as the average power in the numerical value 1 minus the average power in the numerical value 0 divided by the sum of the standard deviation of the noise on 1 and the standard deviation of the noise on 0. It is. The Q factor provides a signal to noise ratio performance index. Previous Q-factor monitoring techniques used either specific high-speed electronics, such as dual decision threshold data recovery electronics, or histogram methods, which were used to detect burst errors. Not very suitable.

  These and other deficiencies of the prior art include, specifically, conventional or all-in-one optical transponder receivers with electronics for processing Forward Error Correction (FEC) encoded data. SUMMARY OF THE INVENTION The system, method and apparatus of the present invention for monitoring defect parameters, such as Q-factor, in an optical communication system are addressed. The error information associated with the FEC correction of the data is used to estimate the bit error rate and Q factor. If an extremely low error condition exists, the identification threshold for the data recovery circuit is adjusted to signal the error to the data stream processed by the FEC electronics.

  A method according to one embodiment of the present invention includes recovering a data signal from an optical signal, processing the recovered signal using forward error correction (FEC), and thereby determining a bit error rate (BER); Determining a parameter of the optical signal defect using the determined BER.

  The teachings of the present invention can be readily understood by considering the following detailed description in conjunction with the accompanying drawings. To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures.

  The subject invention will be described in the context of an optical transmission system basically configured in a particular manner. However, it will be appreciated by those skilled in the art that the present invention may be used to advantage with any type of system that transports data and accompanying FEC information. It is also noted that the subject invention will be basically described in the context of monitoring the Q-factor via BER determinations made using forward error correction coding in the monitored optical signal. It should be noted. However, as will be discussed in more detail below, the present invention generally provides for a bit at either the center (preferred with respect to the Q factor) of the optical signal eye diagram or other parts of the eye diagram. It is possible to apply the use of such forward error correction coding to establish the error rate contour. In each of these cases, the various bit error rate profiles established for the eye diagram of the optical signal provide useful information about the optical performance of the system.

FIG. 1 depicts a high-level block diagram of an optical transmission system using an embodiment of the present invention. Upon identification, the optical transmission system 100 includes a plurality of optical transmission unit ₁₁₀ 1 to 110 _N (collectively optical transmission unit 110) of FIG. 1, the optical multiplexer 115, a plurality of optical amplifying unit ₁₂₀ A to 120 _G (collectively optical amplifying section 120), a plurality of optical add - remove multiplexer (optical Add-Drop multiplexers: OADM ) 130 a and 130 _B (collectively OADM 130), the optical demultiplexer 140, a plurality of optical receiving unit ₁₅₀ 1 _{~150 N} (collectively optical including _OPM) 170 1 _{~170 9} (collectively OPM170): receiver 150), the network management unit 160 and a plurality of optical performance monitor (optical performance monitors.

  Each of the optical transmitters 110 provides an optical output signal having a respective wavelength and having data modulated thereon as well as Forward Error Correction (FEC) information useful for data recovery. . The multiplexer 115 multiplexes the optical signals of the respective wavelengths generated by the transmission unit 110, thereby generating a wavelength division multiplexing (WDM) signal including a plurality (N) of wavelength channels. The WDM signal generated by the multiplexer 115 is propagated to the optical demultiplexer 140 through a plurality of optical network elements such as the optical amplifier 120 and the optical add / drop multiplexer 130. Optical demultiplexer 140 demultiplexes the WDM optical signal and extracts therefrom various wavelength channels initially provided by optical transmitter 110. Each wavelength channel is coupled to a respective optical receiver 150 for further processing.

In the system 100 of FIG. 1, WDM signal following an optical element or first optical amplifying section, produced by the multiplexer 115 (Optical Amplifier: OA) 120 A, the second OA120 _B, the first OADM 130 _A, the third OA 120 _C , the fourth OA 120 _D , the second OADM 130 _B , the fifth OA 120 _E , the sixth OA 120 _F , the seventh OA 120 _G, and the optical demultiplexer 140. It will be appreciated by those skilled in the art that any network path that utilizes one or more optical network elements will benefit from the teachings of the present invention.

  Each of the optical amplifier 120 and the optical add-drop multiplexer 130 is associated with a respective optical performance monitor (OPM) 170. Embodiments of the OPM 170 will be described in further detail below in connection with FIGS. Briefly, each of the OPMs 170 receives a small portion of the optical signal passing therethrough from its respective network element. For example, an optical splitter (not shown) may be associated with each of the optical network elements to divert a small portion of the optical signal passing therethrough (eg, one or two percent of the optical signal power) to its respective OPM 170. There is. OPM 170 selects individual wavelength channels, applies a forward error correction (FEC) configured to recover some of the data propagated through the selected wavelength channels, and remove errors from the received data. To work. An assessment of the bit error rate (BER) is created by determining the number of errors corrected by the FEC processor and knowing the data rate of the extracted wavelength channel data. In some cases, the data slicer's discriminating threshold level (ie, the discriminating threshold used to recover the group of binary zeros and ones from the received data stream) may vary over a range of discriminating threshold levels. The adjustment responsively creates a corresponding bit error rate assessment range. To create a BER contour, the BER assessment can be plotted as a function of the data slicer's discrimination threshold at a fixed time point (eg, the center of the eye diagram). The signal-to-noise ratio (ie, Q-factor) of the wavelength channel can be determined with respect to the shape of the BER contour.

  Thus, OPM 170 operates to determine defect parameters, such as the Q factor, of one or more wavelength channels in the optical signal passing therethrough. By determining the Q factor of each network element in the optical communication system, the source of data degradation experienced by the signal terminating at the receiver goes back to the network element (or communication link) immediately before the OPM indicating the presence of a low Q factor. Can be tracked. It should be noted that the Q factor can be determined for each wavelength channel, a portion of the wavelength channel, or the sample channel that propagates through the system. Further, the cost of the system by associating each OPM with some rather than all of the optical network elements (e.g., every two or three network elements that tend to fail more frequently). It should be noted that it is possible to reduce

  The network management unit 160 has a workstation or a control device including a memory, a processing and an input / output element, which supports the operation of a network element (Network Element: NE) and networking of data. And telecommunications management network (TMN) functions. Techniques used to implement TMN functionality include, for example, overhead bits encoded on a channel, an optical management channel, a dedicated communication link, a packet data network, or any of these. Combinations are also included (not shown).

  The invention can be used in the context of TMN to identify a particular source of signal error or degradation. In one embodiment, each of the plurality of network elements is accompanied by a corresponding OPM 170. By recognizing that the wavelength channels processed by the receiver indicate a high bit error rate, the network manager 160 responds by querying each of the network elements OPM in the optical signal path to their respective BER. Determine the history of the / Q factor / monitor indicator. By determining that the monitor indicator for a particular channel has increased above a threshold from one network element to the next, the network element that initially has the increased BER is defective or in some way The deterioration can be estimated by the network management unit 160. In such instances, it is appropriate to dispatch or reroute repairers (ie, resupply) to avoid such a high level of error experience.

Figure 1 depicts by a third optical amplifying portion 120 _C and a second receiving unit ₁₅₀₂ in the shadow. The shadow process illustrates that the third optical amplifying portion 120 _C is the source of the errors causing ultimately higher BER at receiver unit 150 _2. In this paradigm, BER high reception unit 150 ₂ is noticed by the network management unit 160, which (in the following _order) in response thereto OPM170 9, an inquiry back OPM170 ₈ in so to OPM170 _4. For each query of OPM 170, network manager 160 recovers historical information about changes in the Q factor of each wavelength channel passing through the corresponding optical element. The paradigm case described herein with respect to FIG. 1, the first signs of the third OA120 _C failure or degradation is indicated by the difference in co-channel Q-factor between corresponding OPM170 ₄ and following OPM170 _5. Depending on the arrangement of the OPM (i.e. sampling of the input or output signal of the optical _element), it may be also be that the state of error is detected by comparing the historical data on OPM170 3 (corresponding to the OADM 130 _A) and OPM170 ₄ .

  The TMN may also utilize historical information from the OPM to track the overall performance of the system. Over time, a combination of several factors, each exhibiting a small amount of performance degradation, can impair the overall performance of the system. Changes in system performance can also be caused by irregular repairs, new channels being re-supplied, and other common network changes. By monitoring the evolution of the OPM signal over time, the TMN determines the overall health of the network, thereby determining whether the system is nearing its end of life and whether additional channel loading can be tolerated. Or other network management considerations regarding system health.

  FIG. 2 depicts a high-level block diagram of an optical performance monitor (OPM) suitable for use in the optical transmission system 100 of FIG. Specifically, the OPM 200 of FIG. 2 receives as input signal IN a small portion (eg, 1 or 2%) of the optical signal power propagating therefrom from an optical processing or transmission element within the system 100 of FIG.

  In particular, the optical performance monitor (OPM) 200 of FIG. 2 includes a tuning filter 210, an optical amplifier (OA) 220, an optical tracking filter 230, an optical-electrical (O / E) converter 240, a clock and data recovery circuit 250, It includes a forward error correction (FEC) processor 260 and a controller 270. In some cases, a splitter 275 and a detection unit 280 are also provided. Those skilled in the art will recognize that the choice of components will depend on the particular details of the input signal IN which will normally vary in various communication systems.

An input signal IN, specifically a relatively low power wavelength division multiplexed (WDM) signal, is received by a tuned filter 210. In response to the control signal λ _SEL , the tuning filter 210 selectively filters (ie, blocks) all but one of the wavelength channels included in the WDM input signal IN. The selected wavelength channel λ is amplified by the optical amplifier 220 and coupled to the O / E converter 240. The amount of amplification provided to the selected wavelength channel λ is controlled in some cases by an optical amplifier control signal OAC generated by controller 270. Since the WDM input signal IN is tapped out of the main optical signal path in the optical communication system 100 of FIG. 1, the power level of the WDM input signal IN includes 1% or 2% of the power in the main optical communication channel. It is only possible. Accordingly, the optical amplifier 220 is used to amplify the selected wavelength channel λ to a level suitable for further processing according to the present invention. If a larger tap draw (eg, in the range of 10% to 50%) of the main optical signal is available and the optical loss of the OPM element is sufficiently low, the optical amplifier 220 can be removed from the OPM without changing its function. It is possible to do.

An optional tracking filter 230 is placed between the OA 220 and the O / E converter 240 to filter the optical signals outside of them associated with the selected wavelength channel λ. For example, in the case where the optical amplifier 220 provides a noise component (ie, a spurious component and / or other spectral error) to the selected wavelength channel λ, the power of those frequencies not associated with the selected wavelength channel λ is attenuated. For this purpose, a tracking filter 230 is used. The tracking filter 230 corresponds to the wavelength selection signal λ _sel created by the control unit 270 and previously used by the tuning filter 210. In the case of an optical amplifier 220 that does not exhibit such behavior, the tracking filter 230 can be avoided and the amplified wavelength channel is directly connected to an optical-to-electrical (O / E) converter 240 It is possible to be. One skilled in the art will recognize that other combinations of filters and amplifiers may be used to select a single channel and adjust its power to an appropriate level without changing the function of the OPM. Will do.

  O / E converter 240 converts the optical signal associated with the selected wavelength channel λ into a corresponding electrical signal, which is coupled to clock and data recovery circuit 250.

  In response to the control signal CDRC provided by the controller 270, the clock and data recovery circuit 250 recovers clock information and data (including FEC overhead bits) from the signal provided by the O / E converter 240. Works. The clock information and data recovered by the clock and data recovery circuit 250 are supplied to the FEC processor 260.

  FEC processor 260 is configured to correct errors in the recovered data signal by performing a forward error correction processing operation on the recovered data signal. FEC encoding to data may be provided by digital wrappers that allow monitoring of FEC performance for various data signal protocols. Among such corrections for errors in the data signal, the FEC processor indicates the type and / or amount of error that will be imparted to the data signal in the transmission path between the initial signal source (transmitter 110) and the FEC processor 260. It is possible to supply information. Such information is useful for determining the bit error rate (BER) and other indicators of code error associated with the recovered data signal. In some cases, the recovered data signal DATA is provided to another processing element (not shown) for further processing.

  O / E converter 240, clock and data recovery circuit 250, and forward error correction (FEC) processor 260 together provide functions commonly found in a receiver or transponder. Thus, O / E converter 240, clock and data recovery circuit 250, and FEC processor 260 can be integrated on a common substrate, such as an application specific integrated circuit (ASIC).

  The control unit 270 includes a microprocessor, a memory, and an input / output (I / O) circuit, to give specific examples. The I / O circuit provides an interface between the controller 270 and the various elements communicating with the controller 270. The memory stores a program that, when executed by the processor, performs the steps described herein for the various functions and controls 270. The controller 270 uses the BER data provided by the FEC processor 260 to evaluate the bit error rate of the selected wavelength channel λ and derive a corresponding optical element Q factor therefrom.

  In a first mode of operation, the present invention measures the bit error rate using the optimal identification threshold of the clock and data recovery circuit 250. If an optimal discrimination threshold level is used, the BER from the FEC can be used directly to calculate the Q factor. The first mode of operation is used when there are sufficient errors in the recovered data signal.

  When the Q factor cannot be calculated directly from the BER measurement at the optimal discrimination threshold level, the controller 270 operates in the second mode of operation. Thus, in a second mode of operation, the equivalent Q factor is determined by recording the BER as a function of the discrimination threshold level for some positions on either side of the optimal discrimination threshold level. We now consider this technique in more detail.

ここで等価μ_１、０およびσ_１、０は記号とスペースのデータ・レールの平均および標準偏差であり、Ｄは識別レベルであり、ｅｒｆｃ（ｘ）は次式２で与えられる誤差関数であり、

ここで近似値はｘ＞３についてほぼ正確である。等価Ｑファクタは式１と同様の様式で各々のレールのμとσを使用して計算される。 The equivalent Q factor is measured by recording the BER versus discrimination level at a fixed timing stage, such as below the center of the eye diagram. By fitting this data into an appropriate ideal Gaussian characteristic, the equivalent average and sigma of the symbol group and the space group are determined. The ideal characteristic is given by the following Equation 1 assuming Gaussian noise statistics:

Where the equivalent μ _1,0 and σ _1,0 are the mean and standard deviation of the symbol and space data rails, D is the discrimination level, and erfc (x) is the error function given by ,

Here, the approximation is almost accurate for x> 3. The equivalent Q factor is calculated using μ and σ for each rail in a manner similar to Equation 1.

ここでｘはｌｏｇ（ＢＥＲ）であり、式３は１０^−５〜１０^−１０のＢＥＲの範囲にわたって±０．２％の精度である。これは、データを直線上に乗せるためにＢＥＲ対受信光学パワーの曲線を変形させるように為される計算に類似している。 The calculation of the Q factor is performed as follows. The data is split into two sets with the measured BER dominated by the symbol rail and the space rail. In cases where the SNR is high, the raw data is split at the point of minimum error rate for any measurable BER, or any value of D that results in error-free performance. Each data set is fitted to an ideal curve assuming Gaussian noise statistics to obtain equivalent averages and sigmas for the positive and negative rails. Equation 2 naturally divides into error groups governed by symbol errors and space errors. Once separated, BER is an expression given by a single 1 / 2erfc (·) function. Each set of BER data is passed through the inverse error function, then, the linear regression is performed at decision levels D _i. The equivalents σ _1,0 and μ _1,0 are given by the slope and intercept of the linear regression. To facilitate the calculation, the inverse 1 / 2erfc (•) function is implemented by first taking the log of the BER. Log (erfc (·)) is a smooth one-to-one function, which can be inverted by a number of computational methods, or more simply by using a polynomial fit, as given by Yes,

Where x is log (BER) and Equation 3 is ± 0.2% accurate over a BER range of 10 ⁻⁵ to 10 ⁻¹⁰ . This is similar to a calculation made to modify the BER versus received optical power curve to put the data on a straight line.

FIG. 3 is a graph illustrating a method of determining a Q factor according to an embodiment of the present invention. Specifically, FIG. 3 plots the bit error rate as a function of the identification threshold level. The bit error rate is expressed logarithmically, and the identification threshold level is expressed in amplitude (volts) used by the clock and data recovery circuit 250 for the data slicing function described above. In the graphical representation 300 of FIG. 3, BER is measured at intervals of one second or more and is considered valid if at least five errors have been recorded. The minimum BER was measured at ^10-9 , while the maximum BER was measured at ^10-5 . The equivalent Q factor was determined to be about 8.5. The arrows displayed as an optimum decision point shows the predicted optimum decision threshold level, two vertical sections, labeled as mu ₁ and mu ₀ shows an equivalent mu _i relating rail symbol groups and space groups.

  In particular, scanning the discrimination threshold level between about -0.3 and + 0.3V accumulates errors and leads to the V-shaped curve depicted in FIG. 3 when plotted. The BER approaches a minimum (ie, zero) at the optimum discrimination point and increases as the discrimination threshold level is adjusted, thereby forming a V-shape. By analyzing the V-curve, the Q factor and the bit error rate at the optimal threshold level can be determined. That is, by extrapolating from the plotted curves, the points where they meet provide the optimal discrimination threshold and can be used to determine the Q factor using the methods described above.

  In addition to monitoring the Q-factor described above, FEC error measurements can be used for other signal quality indicators or defect parameters associated with the identification threshold and phase scan. The identification phase point refers to the sample time in the bit slot where the identification is made. If the optimal phase is defined as being in the middle of a bit slot, it is possible to make an error by scanning the phase at the preceding and following sample times within that bit slot. Thus, the eye pattern of the signal is possibly mapped as a function of the number or rate of errors by scanning both the threshold and the phase. Eye patterns are sometimes used to monitor certain defects, such as chromatic dispersion, polarization mode dispersion, or amplified spontaneous emission noise. Other strategies that may be used include adjusting the discrimination level and phase to particular points in the eye pattern and monitoring the number of errors as a function of time. Noise or distortion causes the number of errors at a given point in the eye pattern to vary with signal quality. The identification threshold and phase can likewise be continuously adjusted to maintain a constant number or rate of errors. The required identification threshold or phase will vary with signal quality and will therefore serve as an indicator of signal quality. Those skilled in the art and those informed by the teachings of the present invention will recognize that the number of FEC errors can be used with a variable identification threshold and phase to construct a plurality of accompanying performance indicators. Such uses are contemplated by the present invention.

  In one embodiment of the present invention, rather than creating the entire V-curve as depicted in FIG. 3, only a few points (eg, one point) are monitored to enable fast trend analysis decisions. To That is, in one embodiment, only BERs associated with very few relative threshold positions are processed so that a fast determination of increased BER can be made. By monitoring these one or very few relative threshold positions for a period of time, performance trends are determined with respect to the increase in BER associated with the monitored threshold position (s).

  In one embodiment of the invention, a splitter 275 is used to divert a portion of the amplified wavelength channel to detector 280. In response, detection section 280 determines the power level of the diverted portion, and the power level is communicated to control section 270 as power detection signal Pdet. The detector 280 may include a broadband optical detector such as a pin detector (PIN) or an avalanche photodiode (APD).

In various embodiments of the present invention, the control unit 270 determines that each of a plurality of power levels (as indicated by the power detection signal P _det ) of the selected wavelength channel λ is processed by a transponder circuit. The optical amplification unit control signal OAC is used to controllably amplify the selected wavelength channel λ so as to generate the bit error rate. That is, by selecting each of the plurality of amplification levels and monitoring the selected bit error rate associated with the selected wavelength channel λ for these power levels, for example, at least associated with the selected wavelength channel λ The control unit 270 can generate information useful for determining the Q factor. Further, by selectively processing each of the plurality of wavelength channels in the WDM input signal IN, the control unit 270 can determine the data of the Q factor for each of these data wavelength channels. In this manner, the Q factor of the signal propagating through the communication system 100 of FIG. 1 may be at each of several points or at network elements in the communication system such as the amplifier 120, the optical add-drop multiplexer 130, etc. It can be determined. As previously mentioned, certain imperfections, such as chromatic dispersion, polarization mode dispersion or amplified spontaneous emission noise, also cause the discrimination phase point in the eye pattern to be a non-optimal point (i.e. Adjustment to a point other than the center of the slot).

  The optical transmission system 100 of FIG. 1 can be configured to benefit from the present invention using several structures. For example, in one embodiment of the invention, each of the optical elements 120, 130 is associated with a respective OPM 170. In this embodiment, the OPM continuously adjusts each Q factor so that the network manager 160 can determine which optical element can support high BER reception in response to high BER detection by the receiver 150. Recover. Each of the OPMs 170 is capable of continuously scanning each of the wavelength channels to derive the V-curve information and thus the corresponding Q-factor information. The Q-factor information can be recorded in the OPM as a Q-factor / channel map that is recovered by the network manager 160 in response to the receiver 150 high BER alert. To protect memory, these records may be kept for a limited amount of time.

  In one embodiment of the present invention, at each optical element associated with a respective OPM, the OPM continuously divides the individual wavelength channels to determine Q factor information using one of the techniques described above. Scan. Thereafter, the database is populated, where each wavelength channel is accompanied by a corresponding Q factor, which is periodically updated. In the event of a network alert or other condition, the network manager 160 recovers the Q-factor / channel map from each of the OPMs to identify potential sources of error in the communication system. In a preferred embodiment, upon determining that a high BER condition exists in receiver 150, network manager 160 recovers the Q factor / channel map from the OPM associated with the preceding optics and an error condition at that point. Determine if it exists. The network management unit 160 uses the Q factor / channel map to repeatedly execute the respective Q factor / channel maps (or other defect maps) from the preceding OPM until the OPM accompanying the defective optical element or the channel is determined. ) To recover.

  FIG. 4 depicts a flow diagram of a method according to an embodiment of the present invention. In particular, the method 400 of FIG. 4 is suitable for use, for example, in the control 270 of the OPM 200 depicted in FIG.

  The method 400 enters at step 405 and proceeds to step 410, where data from the selected optical signal is recovered using an optimal identification threshold. At step 415, the recovered data is processed using forward error correction (FEC), thereby deriving the error information.

  At step 420, an inquiry is made as to whether the FEC processing of step 415 has processed or identified any errors. If an error was detected at step 415, at step 435, the Q factor of the selected wavelength channel is determined directly with respect to the optimal identification threshold and error level. That is, control unit 270 operates in the above-described first operation mode to determine the Q factor.

  At step 440, the Q-factor, wavelength channel identification, time and / or other relevant data are stored in respective or unified OPM databases. The next wavelength channel is selected at step 450, after which the method proceeds to step 410.

  If no errors were detected (or were insufficient) in step 420, the data was recovered for each of the plurality of identification threshold levels and processed using FEC to generate a plurality of corresponding BER terms. Is done. As previously mentioned with respect to FIG. 3, one embodiment of the present invention allows for a fast trend analysis function using the monitoring of one or very few identification thresholds. That is, rather than raising the BER data for a plurality of threshold positions sufficient to form a V-curve, a BER contour containing one or very few points is formed instead, and the BER contour is replaced by the previous BER contour. Compared to the contour, it establishes a trend such as an increase in defects at the selection threshold (s), a decrease in defects or a relatively steady defect level.

  At step 435, the Q factor is determined. As previously mentioned, a V-curve created using such data provides a BER contour that can be used to determine the Q factor. That is, the control unit 270 operates in the above-described second operation mode, and thereby determines the Q factor using the BER contour defined by the plurality of BER terms.

  At step 440, the Q-factor, wavelength channel identification, time and / or other relevant data are stored in respective or unified OPM databases. The next wavelength channel is selected at step 450, after which the method proceeds to step 410.

  The method 400 of FIG. 4 may be scanned continuously to form a population, thereby limiting the database to data representing the Q factor associated with each wavelength channel at each of a plurality of times. The amount of time that may be (optionally selectable) is stored. In this manner, for example, subsequent access of the OPM database by the network manager 160 may provide information useful to allow the network manager 160 to determine the source of the reception error.

FIG. 5 is a graph illustrating the defect determination processing method according to the embodiment of the present invention. Specifically, FIG. 5 depicts the bit error rate (BER) plotted as a function of the relative threshold position _VTH . The BER is logarithmic and the relative threshold position is depicted as the offset voltage amplitude used by the clock and data recovery circuit 250 for the data slicing function described above. As discussed previously in connection with FIGS. 3 and 4, forward error correction is determined for each of the plurality of relative threshold positions, and thus the corresponding BER value is determined. The V-curve is then plotted to obtain, for example, the Q factor or other defect parameter of the processed optical signal. It can be seen that the V-curve of FIG. 5 is plotted as a Gaussian noise fit for each of a plurality of relative threshold positions, leading to an intersection of about V _TH = 14 and BER = 1 × 10 ⁻¹⁹ . In the context of obtaining a Q factor, a Q factor is obtained using these parameters in the manner described above. It can be seen that in some cases, a single point may be monitored for fast trend analysis as discussed earlier.

  While various embodiments have been shown and described, which incorporate the teachings of the present invention, those skilled in the art can readily devise many other alternative embodiments that also incorporate these teachings.

1 is a high-level block diagram illustrating an optical transmission system using an embodiment of the present invention. FIG. 2 is a high-level block diagram illustrating an optical performance monitor suitable for use in the system of FIG. 5 is a graph illustrating a processing method of Q factor determination according to the present invention. FIG. 4 is a flow chart illustrating a method according to an embodiment of the present invention. 4 is a graph illustrating a defect determination processing method according to an embodiment of the present invention.

Claims (10)

Recovering the data signal from the optical signal;
Processing the recovered signal using forward error correction (FEC), thereby determining a bit error rate (BER); and using the determined BER to determine parameters of the optical signal defect A method comprising the steps of: Performing the recovery includes selecting each of a plurality of identification thresholds to provide a respective recovery data signal;
Each of the respective recovered data signals is processed using FEC, whereby a respective BER is determined for each identification threshold level, wherein the BER determination establishes a BER contour;
The method of claim 1, wherein parameters of the defect are determined using the BER contour. The optical signal includes a wavelength channel of a plurality of wavelength channels forming a wavelength division multiplexed (WDM) signal;
The step of performing the recovery,
Selecting a wavelength channel from the WDM signal using an optical filter;
The method of claim 1, comprising converting the selected wavelength channel to an electrical signal; and applying an identification threshold level to the electrical signal, thereby recovering the data signal. Performing the recovery includes selecting an optimal identification threshold level to provide a respective recovery data signal;
The respective recovered data signals are processed using FEC, whereby a respective BER is determined;
The method of claim 1, wherein parameters of the defect are determined using the BER.   The method of claim 1, wherein the parameter of the defect includes a Q factor.   The method of claim 1, wherein the parameter of the defect comprises at least one of chromatic dispersion, polarization mode dispersion and amplified spontaneous emission noise. An optical performance monitor (OPM),
A tunable filter for selecting a wavelength channel from a plurality of wavelength channels forming a wavelength division multiplexed (WDM) optical signal; and recovering data from the selected optical signal and performing forward error correction (FEC). A receiver for processing the recovered data and thereby determining a bit error rate (BER), and a controller for determining a parameter of the optical signal defect using the determined BER. OPM including. The receiving unit,
An optical-to-electrical converter for converting the optical signal into an electric signal,
8. A data recovery circuit for recovering a stream of data bits according to an identification threshold level, and an FEC processor for identifying errors in the stream of data bits and thereby determining the BER. OPM according to the above. A plurality of network elements for propagating an optical signal carrying data from a transmitter to a receiver, wherein the receiver monitors a bit error rate (BER) of the recovered data;
A plurality of optical performance monitors (OPM) for monitoring at least one defect parameter of the data-bearing optical signal immediately before at least some of the plurality of network elements; and identifying degraded network elements by examining a Q factor. A system that includes a network manager responsive to the BER representing the degraded network element at the receiver. The system of claim 9, wherein the defect parameters include at least one of a Q factor, chromatic dispersion, polarization mode dispersion, and amplified spontaneous emission noise.

Images